the risks of commercial real estate lendingpdf.usaid.gov/pdf_docs/pnacj140.pdf · impact of the...

THE RISKS OFCOMMERCIAL REAL

ESTATE LENDING

FINAL REPORT

Prepared for Prepared by

East European Regional Urban Sector Assistance Project withProject 180-0034U.S. Agency for International Development, ENI/EEUD/UDH Achim DübelContract No. EPE-C-00-95-001100-00, RFS No. 624

Michael J. LeaCardiff Consulting Services

Jacek ºaszekCracow Real Estate Institute

contributions from

Loïc ChiquierWilliam HandorfAbt Associates

under subcontract to

The Urban Institute

THE URBAN INSTITUTE2100 M Street, NWWashington, DC 20037(202) 833-7200 December 1997www.urban.org UI Project 06610-624

TABLE OF CONTENTS

EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

BANKING CRISES AND COMMERCIAL MORTGAGE LENDING — COMMONCONJECTURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

COMMERCIAL MORTGAGE CREDIT RISK — DETERMINING FACTORS . . . . . . . . 3Commercial Property Price Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Commercial Mortgage Market Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Impact of the Legal System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

UNDERWRITING COMMERCIAL MORTGAGE LENDING RISK . . . . . . . . . . . . . . . 14Loan-to-Value Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15Debt Service Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Valuation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

COMBINING THE UNDERWRITING FACTORS: RECENT U.S. EXPERIENCE . . . 25Regulatory Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25Regulatory Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

COMMERCIAL REAL ESTATE LENDING IN POLAND . . . . . . . . . . . . . . . . . . . . . . . 31Overview of Real Estate Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Commercial Real Estate Lending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

LESSONS FOR REGULATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Promote Stable Property Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36Improve the Information Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36Strengthen the Legal System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Improve Bank Regulation and Supervision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

DATA ANNEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

EXECUTIVE SUMMARY

Market-based commercial real estate lending is quite new in Poland, havingdeveloped only in the past three years. As of yet there are only a few players in themarket and there are no standardized underwriting procedures, documentation ordetailed regulations. The National Bank of Poland (NBP) has primary responsibility forthe safety and soundness of the banking system. As Polish banks enter thecommercial mortgage market it is essential that the NBP develop appropriateguidelines to monitor and regulate the risk of this sector to the banking system.Likewise, following passage of the Mortgage Bank Act, it will be necessary to developguidelines for the commercial real estate lending activities of mortgage banks and forthe use of commercial real estate loans as collateral for mortgage bonds. This is ofparticular concern for the Ministry of Finance which has been charged with draftingthese guidelines.

The purpose of this study is to provide the NBP and the Ministry of Finance witha comprehensive overview of the risks of commercial real estate lending and how theserisks are managed, regulated and supervised in selected European countries and theUnited States. The risks of commercial real estate lending depend critically on the typeof lending, in particular whether loans are for property development or standinginvestment projects. It is the former more than the latter that is responsible for theenormous losses suffered by lenders in a number of Western countries during the1980s and 1990s. Furthermore, the risk of commercial real estate lending depends onthe legal standing of the lender (i.e., whether the lender has the right to foreclose on adefaulted loan and repossess the property within a reasonable time period and forreasonable cost) and the lessor (i.e., whether long-term lease contracts can be offeredand enforced).

Lender underwriting is the key risk control variable. The focus of underwritinghas traditionally been on the loan-to-value ratio and the debt service coverage ratio.Property valuation plays a key role in this process and the different approaches tovaluation are highlighted in this study. Particular emphasis is given to the Germanconcept of mortgageable value which is a regulatory guideline for valuation ofproperties collateralizing loans used as security for mortgage bonds. The recent U.S.experience is quite instructive as well. As the funding of commercial real estate hasshifted away from traditional intermediaries such as commercial banks and savings andloans and towards securitization, the rating agencies have become more prominent inthe risk review process. They have developed detailed, quantitative rankings of therisks of commercial real estate backed securities that provide useful guidelines forregulators.

The Risks of Commercial Real Estate Lendingiv

The last section of the study points to a number of lessons from the Europeanand U.S. experience for Poland. These include:

• The promulgation of policies leading to stable property markets.

• The importance of developing property and loan databases.

• The need to create a creditor-friendly legal system, in particular by improving thelien position of mortgage lenders.

• The harmonization of risk-based capital treatment of commercial and residentialmortgage loans at a 100 percent risk weight at least until a body of Polish-specificexperience and information can be developed and more differentiated guidelinescan be adopted.

• The necessity to discriminate the regulatory treatment of long-term standinginvestment loans and development loans and the importance of providing incentivesfor loans with high investor equity (lower loan-to-value ratio) and proper debtservice coverage guidelines.

• The importance of developing banking supervision expertise and regulatoryconsistency in this important area of bank lending in Poland.

THE RISKS OF COMMERCIAL REAL ESTATE LENDING

BANKING CRISES AND COMMERCIAL MORTGAGE LENDING −− COMMONCONJECTURES

Credit losses associated with a surge of mortgage lending activity during the late1980’s and early 1990’s have been a prominent, in some cases decisive, factor inbanking crises that characterized the past decade1. Declines of both commercial andresidential property prices were important factors causing banking problems the Japan,the United States, the United Kingdom, France, Scandinavia, as well as more recentlyin a number of Latin American and South East Asian countries. The massive financialcosts for the financial sector, and most notably for governments standing ready torescue ailing financial institutions, triggered broad discussions about the causes of therecent crises and the appropriate regulatory response. Typically, the literature hasfocused on the following combination of macroeconomic factors:

• The effects of internal financial sector deregulation, which changed the rules of thegame in many countries during the 1970’s and 1980’s. Lifting restrictions on depositraising spurred the growth ambitions of many financial institutions in developedcountries in the 1980s. Furthermore, the removal of many traditional investmentrestrictions for lenders and consequent portfolio diversification taxed their capacitiesto manage their new risk exposures properly.

• The effects of financial globalization and capital account liberalization, allowingmobile capital to flow with negligible transaction costs to hot investment spots.Special emphasis is typically made on the huge net external creditor position ofJapan that had built up by the end of the 1980s. This helped to generate strongand volatile cross-border capital flows and exacerbated the cycles first in domesticJapanese, and later in other local property markets.

• The effects of stop-and-go monetary policy, which generated much greater financialmarket volatility than before under these new structural conditions. By the end of1988, short-term interest rates had dropped to the lowest levels in the majordeveloped countries for nearly a decade, only to rise again by 3 percent in the U.S.and 5 percent in Germany over the next 18 months. Assets valued with these short-term rates suffered correspondingly high losses.

1 For overviews over recent banking system failures in developing and developed countries, see Honohan1997) and Goldstein and Turner (1996). For commercial mortgage aspects of the lending crisis worldwide seeSchinasi and Hargraves (1995) and Renaud (1995).

The Risks of Commercial Real Estate Lending2

• The cyclically repeating overexposure of the banking industry to key “high-risk”asset classes, such as commercial property lending, as well as a lack of generalconstraints on such investment. The key argument is that commercial propertymarkets react differently from other markets to shocks and, by their very nature,create a high-risk asset category.

The macroeconomic factors described in the first three points above can only benecessary conditions for property-related lending crises (that is, none are in a strictsense causal); furthermore, most of the factors are beyond the reach of regulators andeven domestic economic policy makers. Therefore, many regulators concentrated theirconclusions on the last point, leading partly to extremely restrictive approaches thatwere designed to discourage lenders from investing in commercial property. Examplesof strong regulatory reaction included noted actors, such as the Bank of England2 andthe Bank of Canada3. In addition, many emerging economies (Singapore, Philippines,Malaysia) have periodically restricted and relaxed commercial real estate lendingguidelines, often creating a regulatory property price cycle4.

Given these conclusions with respect to the character of commercial propertymarkets and lending behavior, the question becomes identification of suitable, lesscyclical, regulatory approaches. This approach implies concentrating on themicroeconomic factors of commercial mortgage lending that may control commercialreal estate lending risk, including:

• The structure of the commercial mortgage markets, in particular risk behaviorsdiffering by property types, investor types, funding/lending techniques, as well asthe nature of the price cycle.

• The legal system determining the relations between lender and property investor, aswell as between property investor and commercial tenant.

• The nature of bank lending itself, with risk features differing among other things bylender specialization and diversification, valuation techniques, credit underwriting,foreclosure and property management standards.

2 Systematic discouragement of lenders from real estate lending.

3 Limiting bank real estate investment and lending to 70 percent of regulatory capital.

4 Interestingly, the most seriously affected economy, Japan, has not introduced any harsh regulatorymeasures for commercial mortgage lending; similarly in France. A review of a number of cases worldwidereveals no systematic relationship between the onset of commercial mortgage lending crisis and regulatoryaction ex-post.

The Risks of Commercial Real Estate Lending 3

While the importance of microeconomic factors regarding the financial sector isstressed by most authors5, special conclusions arising from the nature of commercialproperty are often absent. This report will summarize some key points, elaborating oninternational empirical evidence brought up in the empirica study by Dübel, Pfeiffer et.al. (1994). 6

COMMERCIAL MORTGAGE CREDIT RISK −− DETERMINING FACTORS

Commercial Property Price Risk

The relative price volatility of commercial and residential real estate is shown inFigure 1 and Figure 2. These data convey the impression that commercial propertyprices are intrinsically more cyclical, for quite a number of countries, than residentialproperty prices. A problem is that most commercial property price data, including thatdisplayed, stem from surveys of real estate agents which typically report only thevolatile, high-price, top-location-top-quality end of the market. Also, office price dataare often taken as a proxy for the entire commercial property market (which alsoincludes retail, hotel, industrial, and other uses) for which they may not berepresentative. In contrast, residential property prices are typically covered by publicspecial surveys, and reported in detail by national statistical offices.

The empirica study shows that a generalized view of an intrinsically higher pricerisk in commercial uses of real estate needs correction for the following reasons:

• The cyclical behavior of high-end and mean values for office rents or capital valuesdiffers strongly, as Figure 11 in the annex shows for Germany. These observationshold for Italy, the UK and other surveyed countries (note that residential propertyprice data published in statistics typically refer to mean or median concepts).

• High commercial mortgage price cyclicality is a phenomenon of “hot spots,” that isthey tend to be geographically concentrated. For example, the study shows that in 9out of 13 UK regions, volatility of house prices exceed the volatility of office capital

5 Honohan (1997) speaks of the „three failures syndrome“: macroeconomic epidemics, poor managementand microeconomic deficiencies, and endemic crisis in a government-permeated banking system.

6 This section is based on a study undertaken by empirica for the Verband Deutscher Hypothekenbankenconcerning Article 11(4) of the EU-Solvency Directive in 1994/1995. The article granted a temporary derogationfrom the general 8 percent Basle equity rules to a 4 percent level for certain commercial mortgage loans in thecountries of Denmark, Greece and Germany until 1995. The central point raised in the derogation was whetherfirst mortgages within a 60 percent loan-to-value range were sufficiently effective in cushioning the impact ofprice shocks on the underlying collateral markets to the bank’s asset quality, or whether additional risk bufferswould prove necessary in order to create a low-risk/low-cost commercial mortgage loan benchmark.


values during 1981-1993. Only in Scotland and London were office capital valuesmore volatile (Figure 12; likewise for Paris, Figure 13) and suggest that there canbe no generally valid inference on higher price risk for commercial real estate assuch.

• Depending on which activity is financed – long-term standing investment or propertydevelopment (see below) – different price indices are the appropriate benchmarks.Standard statistics for residential rental housing typically make this importantdistinction.

• Even in hot spots, there are very different kinds of cycles: a London-type cycle,where short-term production incentives and disincentives generate a confined short-term cycle (that leaves the long-term investor basically unaffected); and a Paris-typecycle, where structural adjustments (large increase in the supply of developableland due to planning change) created a much stronger long-term price adjustmentprocess.

• Finally it should not be overlooked that commercial property markets comprise verydifferent uses, with some premises offering the investor a conversion option, forexample from office into retail or housing, and vice versa.


Figure 1: BIS Data - Commercial Property Prices

0

50

100

150

200

250

300

350

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

Inde

x 19

86 =

100

Madrid

Sydney

Helsinki

Tokyo

Stockholm

London

Milan

Frankfurt

Paris

Oslo

Source: empirica (Bank for International Settlements)

Figure 2: BIS Data - Residential Property Prices

0

50

100

150

200

250

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

Australia

Finland

Japan

Sweden

United Kingdom

Italy

Germany

France

Norw ay

Canada

USA

Source: empirica (Bank for International Settlements)


Commercial Mortgage Market Structure

Sources of Funds

Commercial real estate investment is funded from a mixture of debt and equity.At the end of 1996 in the U.S. there was over $738 billion in non-residential, non-farmcommercial real estate debt outstanding (Federal Reserve) representing approximately9.7 percent of GDP.7 This contrasts with $3.9 trillion of residential (defined as one-to-four family) mortgage debt outstanding (52 percent of GDP). Commercial banks arethe largest supplier of mortgage debt, holding approximately 50 percent at the end of1996. (Figure 3). Life insurance companies are the second largest holder at 22 percent.The fastest growing segment of the commercial real estate debt market is securitizedpools which grew from $15 billion in 1992 to $63.5 billion in 1996.

Figure 3: U.S. Commercial Real Estate Debt Outstanding

U.S. Commercial Mortgage Debt Outstanding$738.3 billion

1996 (Federal Reserve)

Commercial Banks50%

Thrift Institutions8%

Life Insurance Cos.22%

Federal Agencies1%

Mortgage Pools8%

Individuals/Others11%

Source: Federal Reserve

7 Multifamily mortgage debt is treated as residential. There was $310.4 billion of multifamily (5+ dwellingunits) mortgage debt outstanding at the end of 1996.


There are no comprehensive data on equity investment in commercial real estatein the U.S. As of 1994, real estate holdings of both life insurance and pension fundswere in excess of $60 billion (Clauterie and Sirmans, 1995). Life insurance companiesheld between 3 and 4 percent of their assets in real estate and pension funds heldbetween 4 and 5 percent. There has been a growing trend towards public ownership ofreal estate in the form of Real Estate Limited Partnerships and Real Estate InvestmentTrusts (REITs). Tax law changes have had a large influence on the ownership of realestate in recent years. The Tax Reform Act of 1981 provided significant incentives forindividual ownership through limited partnerships. Most of these incentives were takenaway in the Tax Reform Act of 1986 (with a predictable negative effect on real estatevalues).

According to data collected by the European Mortgage Federation (EMF), therewas more than $3.4 trillion of mortgage debt outstanding in Europe as of 1996. This isan underestimate as they do not report commercial mortgage debt outstanding forFrance, Finland and the UK. Table 1 shows the strong differences in both residentialand commercial mortgage market penetration among European countries, suggestingthe existence of fundamental differences.

Table 1 Mortgage Lending in Europe

CommercialMortgage Debt/GDP

(%)

Mortgage DebtOutstanding/GDP

(%)

Gross ResidentialLending as% of GDP

Gross CommercialLending as% of GDP

Norway 30.72 60.08 N/A N/A

Denmark 22.34 55.76 13.79 11.09

Netherlands 22.82 55.73 14.07 5.01

United Kingdom N/A 54.44 9.70 N/A

Sweden 18.73 51.14 1.44 N/A

Germany 9.89 42.21 3.30 N/A

Belgium 3.37 22.81 5.06 0.29

Ireland 4.07 21.45 4.66 0.00

France N/A 20.53 2.62 0.00

Portugal 4.84 20.32 5.98 0.57

Spain 8.19 19.46 4.80 2.78

Italy 3.83 7.30 0.43 0.19

Source: European Mortgage Federation, authors’ calculations

When conducting cross-country comparisons, one has to keep in mind thatmortgages or any other type of financing of commercial property is provided underdiverse institutional and legal environments which influence the competitive position of


mortgage loans. These differences are often not the result of market forces, but ratherof different legal and institutional frameworks. The empirica analysis shows that thestructure of long-term finance and the significance of mortgage lending in each countryis strongly influenced by the following factors:

• The presence or absence of tax incentives for private investors in real estate.

• The presence of long-term institutional investors, such as pension funds andinsurance companies.

• Core legal characteristics of the mortgage loan, such as performance in bankruptcyand foreclosure procedures.

The central question of leverage, i.e., the proportion of equity vs. debt financingof the property, depends largely on the legal position of the creditor (see below) andthe resulting regulatory constraints on large institutional investors.

Table 2 displays estimates of the importance of institutional investors as equityinvestors in property in Europe. Clearly, the significance of institutional investors asequity investors in property is much higher in the UK and the Netherlands than, forinstance, in the mortgage bank countries of Germany and Denmark.

Table 2 Estimated Real Estate Holdings of Financial Institutions in Europe

1994 Pension Fund Assets Life Insurance Assets Approximate Total

CountryAssets % of

GDPbReal Estate% of Assets

Assets % ofGDPc

Real Estate% of Assets

Real Estate% of GDP

Real Estate% of Assets

Denmark 20.1 9.0 49.0 3.0 3.3 4.7

Germany 5.8 11.0 20.5 5.0 1.7 6.3

Spain 2.2 1.0 4.2 10.0 0.4 6.9

France 3.4 7.0 30.0 8.0 2.6 7.9

Italy 1.2 10.0 3.7 12.0 0.6 11.5

Netherlands 88.5 10.0 45.6 6.0 11.6 8.6

Sweden 33a 8.0 36.9 7.0 5.2 7.5

UnitedKingdom

79.4 6.0 67.1 9.0 10.8 7.4

Notes a 1991 b 1993 c 1995 Source: Commission of the European Union (1997), author’s calculations

In Germany, with its extensive pay-as-you go social security system, privatepension funds are weak in comparison to those in Britain. At the same time, portfolio


structure analyses show that German insurance companies and pension funds preferindirect investment via bonds and loans, while British pension funds prefer equityinvestment. These differences are, among other things, the consequence of statutoryregulations. The fact that in Germany, as in Denmark, a low risk loan type is defined bya specific set of regulations makes it easier for pension funds to buy bonds or givedirect mortgage loans, while for British funds, equity investment has been traditionallymore attractive. In France the Credit Bail (leasing) system was developed for tax andcredit risk reasons.8

Property Development Lending or Standing Investment Loans

The structure of commercial mortgage markets is often misunderstood; as notedfrequently in the discussion about the risk of commercial mortgage lending, nodifferentiation is made between development loans (“property lending”) and long-termloans for standing investments. These two loan categories represent very differentconcepts: development loans, unlike standing investment loans, bear the completionrisk, the construction time risk, and a much higher underlying price risk. Furthermore,the volume of new property lending is extremely cyclical, reacting to, or inducing, newinvestment in commercial real estate. By contrast, outstanding standing investmentfinance is sluggish in response to new investment. Figure 4, comparing data fromGermany (high share of standing investment finance) and Britain (high share ofproperty lending), demonstrates the point.

8 In a leasing arrangement the bank owns the property and leases back to the user. The lease contractcan last for 15 years and the lessee has an option to purchase the building after 5 years.


Figure 4: Comparison United Kingdom/Germany - Non-residential Mortgage vs. PropertyLending

-10

-5

0

5

10

15

20

25

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

billi

on D

M

-4

-2

0

2

4

6

8

10

12

billion £

German Banks: Change inOutstanding Non-residentialMortgage Loans

United Kingdom Banks: Net Flowof Funds in Property Loans.

Note: Germany after 1990 not comparable Source: empirica

Development loans are much riskier than long-term standing investmentcommercial mortgage loans. Figure 5 shows data reported in the Report of Conditionsof the U.S. Federal Deposit Insurance Corporation (FDIC), covering loans bycommercial banks during the period of 1991-1994. There are limited differences indelinquency (default) rates between single-family loans, commercial loans (non-residential/non-farm), and multi-family loans; on the other hand, however, developmentand construction loans show twice the delinquency rates of commercial loans.9

9 The definition of development loans is not easy, as in many cases, bridging loans during the constructionperiod can be seen as the first installment of a long-term loan.


Figure 5: U.S.: Real Estate Loan Delinquency Rates as a Percent of Average Loans,Seasonally Adjusted Quarterly Averages (All insured Chartered Commercial Banks)

0

2

4

6

8

10

12

14

16

18

20

I/91 II III IV I/92 II III IV I/93 II III IV I/94 II

Residential

Multifamily

Nonfarm/Nonresidential

Construction and LandDevelopment

Source: FDIC Report of Condition

The difficulties in differentiation also arise from corresponding banking statistics,in which loans to the real estate sector are treated as one category regardless of theirfunction. Lending to the property sector in Britain, for instance, nearly exclusively refersto loans given to property development companies that typically operate under thehighly volatile conditions of a narrow domestic property market. It is no surprisetherefore, that research in Britain finds much higher default rates in comparison toresidential lending.10

It would be a misunderstanding, however, and impossible in the British datacontext, to speculate about the relative default risk of commercial real estate loans,including long-term standing investments, given the institutional context in which thereis hardly any long-term lending.

10 E.P.Davis (1993) finds, for instance, in a long-term analysis of a large British clearing bank over theperiod between 1976 and 1991, mean write-offs of 0,05 percent for loans for house purchases and 0.60 percentfor lending to property companies.


Impact of the Legal System

Legal Position of the Creditor

There are two key issues in any credit risk analysis: the likelihood of default andthe loss the lender incurs in the case of default.

The amount of loss in case of default is strongly influenced by the legal positionof the creditor. In countries with strong protection of lenders’ rights to collateral in theevent of default, such as in the mortgage bank countries of Denmark, Austria, orGermany, or common law countries such as Britain and the U.S., the mortgage is anabsolute claim. Only property taxes and fees for public infrastructure have to be paidfirst out of property sales. The lending institution has direct access to a property andcan use the receipts from sale to cover the outstanding loan balance plus interest.

This situation differs in countries with borrower-friendly, or complicated,bankruptcy procedures. These countries typically have undersized commercialmortgage sectors and have developed alternatives to commercial mortgage loans forthe capital market. The classic case is France, where leasing dominates commercialproperty finance in part due to the weaknesses of the legal position of the mortgage.Bankruptcy procedures are typically avoided by mortgage lenders, as these givepreference to other claims (unpaid salaries, all unpaid taxes), forcing lenders to enteran amicable solution vis-à-vis a magistrates court. This situation has compoundedFrench lenders’ losses during the extended property cycle.

The ability to realize the value of the mortgage as collateral has been a keycontributor to the varying market penetration of commercial mortgage lending and alsoto loan loss performance in different countries. In those countries in which mortgageloans have an absolute priority over other claims in the bankruptcy procedure, lendinghas proven to be an attractive alternative to equity investment for institutional investorsas risks are lower and less knowledge is necessary to participate indirectly in the realestate markets. In particular, where a low risk type of loan is defined by strictregulations, banks and their clients can invest capital at low transaction costs and withlow information costs. These facts have been acknowledged by regulators which allowinstitutions in some countries to hold large mortgage loan books, and in others whichare much more restrictive.

Legal Position of the Lessor

Whether projected rent revenues may cover the debt service of a commercialmortgage loan depends crucially on the form of commercial rent legislation. In someEuropean countries (for example, France and Belgium), rental legislation allows foruniversal contract termination options for commercial tenants, limiting the maximumperiod over which rental revenues can be effectively estimated. In Japan, one of the


key factors in the sharp office price decline was the fact that commercial tenants areuniversally allowed to give notice to landlords every 6 months.

Countries with strong lessor rights (for example, United Kingdom and Germany)experienced less price cyclicality and default problems with rented properties. In theUnited Kingdom, the data show that rented commercial properties were a powerfulhedge against the property price cycle; unrented stock, in turn, was hit hardest.

Unilateral tenant options, combined with uncertainty over timing and frequencyof their exercise, make the anticipation of rent revenues difficult; thus they may causeprice cycles to intensify. The fundamentally different price cycles in Paris and Tokyomarkets in comparison to London are a clear indication of this.

Table 3 displays basic information with respect to the commercial rent legislationin seven EU member states. In Germany, Spain and the Netherlands the typical term ofrental contracts is five to ten years; In Britain it is still ten to fifteen years. In thepresence of long contract terms, there is generally no legal right for the tenant toachieve extension.

In Belgium and France commercial rent contracts have a typical maturity of nineyears. It is important to note, however, that in Belgium both parties have the right toterminate the contract every three years; in France only the tenant has such a right.Tenants have also the right to demand extensions in both countries. In Italy the typicalcontract lasts six years with a legal option of extension. Only in Britain may rentalchanges be implemented on the basis of market rents. In other countries the consumerprice index or construction cost index forms the basis of rent increases.


Table 3 Commercial Lease Legislation In Selected EU Countries

CountryTypical ContractDuration/Years

Tenant Option onExtension?

Adjustment ofConditions Frequency

Belgium 9 (3/6) yes CPI Contractuallydetermined

Germany 5 to 10 no CPI Contractuallydetermined

Spain 5 to 10 no CPI N/A

France 9 (3/6) yes CCI Annual

United Kingdom 10 to 15/25 yes Market Rent Freelynegotiable,

typically every 5years

Italy 6 yes CPI Typically every 5years

Netherlands 5 to 10 no CPI Annual

Notes CCI = Construction Cost Index CPI = Consumer Price Index Source: IFO-Institut, authors’ research

In the U.S., office leases are typically 5 to 10 years with 7 years being mostcommon. Retail property leases are frequently longer. Leases for credit anchors (large,well known stores that draw other vendors and customers to a center) can be 20 yearsor longer. Smaller non-credit anchors are around 10 years and other lessees aretypically 5 years. Warehouse leases are shorter (typically 5 years). Leases arefrequently indexed to the CPI.

Restrictive rent legislation can be very detrimental for commercial landlords,increasing the default risks of commercial mortgage lending by restricting the ability toraise rents in line with expenses. Mandatory unilateral tenant options in the commercialsector, rooted in the idea to protect small enterprises, can prove highly detrimental tofinancial stability goals.

UNDERWRITING COMMERCIAL MORTGAGE LENDING RISK

The principal determinants of mortgage default are the willingness and the abilityof the borrower to pay. The willingness to pay is based on the amount of equity theborrower has in the property as determined by the current loan-to-value ratio and theability to pay is based on the ability of the property to produce sufficient cash flow toservice the debt.


Loan-to-Value Ratio

It has been shown that the loan-to-value ratio (LTV) by far dominates otherunderwriting criteria as well as property characteristics as a risk buffer; that is,conservative (low) LTVs will result in both a lower default risk and lower losses in caseof a default. This is supported by economic theory which considers loan default as aput option of the borrower, giving him the opportunity to abandon the project if hisequity becomes depleted. This typically will happen if property prices drop strongly.

Figure 6 shows the power of loan-to-value restrictions for the case of a largeresidential mortgage loan pool (725,000 loans) which was analyzed by Van Order(1990). He developed an econometric model designed to capture the impact of loan-to-value ratio and environmental factors (approximated by the loan origination year(interest rates, borrower characteristics). The result of the analysis is that conservativeunderwriting clearly dominates other characteristics. While, for instance, loansoriginated in the high interest-rate environment of 1981 are still subject to a higherdefault rate, sufficiently high equity in the project reduces the default likelihood tominimal levels.


Figure 6: Cumulative Default Probability of Single Residential Mortgages Cohorts within 10 years of Origination

0%

2%

4%

6%

8%

10%

12%

14%

16%

80% or less 81 to 90% 91 to 94% 95%

Underwri t ing loan-to-value

1981

1982

1980

1983

1979

1978

1977

1976

Source: Van Order (1990)

Similarly, loan-to-value restrictions provide substantial support for the creditquality of commercial mortgage loans. Research by Vandell et al. (1993) has found asimilar result with the ratio of the market loan-to-value ratio by far the most significantexplanatory variable in an analysis of defaults on commercial real estate loans made bylife insurance companies.

The empirica study examined the role of underwriting in the loss experience ofGerman Mortgage Banks. Figure 7 presents a time series of loss ratios for 60 percentLTV first mortgages from 1977 to 1991. The data represent write-off balancestatements for the sample of 26 mortgage banks which are divided through outstandingloans in the categories of residential and commercial mortgage collateral. Loss ratiososcillate around a very low level of 4.6 basis points (a basis point is 1/100 of apercentage point) for residential mortgage loans (including both owner-occupiedhousing and multi-family housing) and 5.4 basis points for commercial mortgageloans.11 These data show no systematically different loss behavior between residentialand commercial loans, reflecting the conservative underwriting (maximum loan-to-value

11 As of 1993, offices made up for 38.8 percent of total commercial lending of the mortgage banks, andmulti-purpose buildings for another 35 percent.


ratio of 60 percent) and the fact that German mortgage loans in general feature a highshare of standing (fully leased) investment finance loans.

Figure 7: Loan Loss Rates of German Mortgage Banks 1977-1991 - Residential vs. Commercial First Mortgages (60 percent Loan-to-Value)

0

2

4

6

8

10

12

14

16

1977 1979 1981 1983 1985 1987 1989 1991

Residential Mortgages

Commercial Mortgages

Note: Loss rates are indicated in basis points, i.e., hundredths of a percentage point Source: empirica

In the empirica study, a survey among 4 mortgage banks sampled 1,900disaggregate loan files over the period of 1980-1994. The estimates from aneconometric model supported the conjecture of the importance of the underwriting loan-to-value as a risk buffer: the model estimated that for every 20 percent increase in theloan-to-value ratio the default probability of an office loan would rise by 43 percent. Anumber of other analyses with disaggregate data sets support these findings (e.g.,Vandell, et al., 1993). Key to all the analyses is that the loan-to-value ratio dominatesall other loan characteristics, regardless of whether the use of the property isresidential or commercial.

During most European price cycles of the late 1980s to the early 1990s a firstmortgage position up to 60 percent of the property value would have been sufficient inorder to insulate a diversified lender’s portfolio sufficiently from any losses, as theremaining collateral would have suffered to cover the lender’s claim. However, thisstatement does not hold for some single-vintage portfolios (e.g., underwriting in 1989),concentrated on very volatile markets (e.g., London). Additional safety features aretherefore needed.


Debt Service Coverage

The debt-service coverage ratio is defined as the ratio of net operating income tothe mortgage payment. As shown in Table 4, a project generating a gross rent of$600,000 in year 1 with a vacancy rate of 5 percent and an expense ratio of 25 percentof gross rent would generate net operating income of $420,000. If the owner obtainedan 80 percent LTV mortgage based on a project valuation of $2,800,000 at a 12percent fixed rate for 20 years the annual debt service payment would be $299,888.The initial debt-service coverage ratio (DCR) would be 1.4 ($420,000/$299,888) whichin this example would grow to 1.84 in 5 years assuming 7 percent annual increase innet rents. The risk in focusing on the initial DCR is that rents may not increase at therate forecast (e.g., if they are not indexed), vacancies may rise, or expenses may riseat a faster rate than assumed. Also, if the borrower takes out a variable rate loan themortgage payment may not remain constant.

Table 4 Project Cash Flows

Year 1 Year 2 Year 3 Year 4 Year 5

Gross Rent $600,000 $642,000 $686,940 $735,026 $786,478

Vacancy $ 30,000 $ 32,100 $ 34,347 $ 36,751 $ 39,324

Net Rent $570,000 $609,900 $652,593 $698,275 $747,154

Operating Expenses $150,000 $160,500 $171,735 $183,756 $196,619

Net Operating Income $420,000 $449,400 $480,858 $514,518 $550,534

Mortgage Payment $299,888 $299,888 $299,888 $299,888 $299,888

Before Tax Cash Flow $120,112 $149,512 $180,970 $214,630 $250,646

Debt Coverage 1.40 1.50 1.60 1.72 1.84

Break-Even Occupancy 0.75 0.72 0.69 0.66 0.63

Operating Expense 0.25 0.25 0.25 0.25 0.25

Where valuation techniques are not sufficiently developed, or price cycles areextremely strong, a debt coverage requirement can be implemented in combination withor substitution for the loan-to-value ratio restriction. An adequate debt service coverageratio is technically easy to manage, as lenders will typically collect information aboutrent contracts. This information could also be used by bank supervisors.

For standing investment finance, the stability of a debt service indicator dependsmuch on contractual provisions of the rental agreements, or equivalently pre-leases, aswell as the financial conditions of the loan, which must be sufficiently predictable overthe loan time period.


Figure 8: 1989 London Market Analysis: Percentage of Office Properties Acquired in 1989 Yielding Below X percent in the Subsequent Years

0

10

20

30

40

50

60

70

80

90

1990 1991 1992 1993

<12%

<10%

<8%

<0%

Source: empirica

To show the power of the debt-service coverage ratio as a risk cushion theempirica study analyzed the London office market. For simulation purposes, adisaggregate bank property loan books was constructed with data generated by IPD's12

London Office Market survey. The study concentrated on the worst case: officebuildings built or bought at the top of the price cycle in 1989. For this vintage, even a60 percent loan-to-value restriction would not have been sufficient to protect theaverage lender from a loss in the event of being forced to liquidate the property at thebottom of the cycle, in 199313.

Figure 8 shows the simulation results. It indicates the percentage of officeprojects in the database which generated net rents (yields) at a specific return in theyears following 1989, the peak of the cycle. A small but growing percentage generated

12 IPD: Investment Property Databank, is an information service that provides for a comprehensivedatabase of financially relevant characteristics of British commercial properties. It has been founded andsupported by key investors in the property market (mainly pension funds), preforming thereby a public goodinformation function for all investors in the market.

13 The argument has, however, to be seen against the background of the open market valuation techniquethat is typically also applied to lending operations in Britain (see below).


nearly zero revenues – that is, vacant buildings. Thirty to forty percent had insufficientreturns for their investors (below 8 percent, the level likely to cover the debt service onan assumed mortgage), clearly an indication that the market was significantlyoverpriced during 1989. The key finding is that as property prices dropped precipitouslyin 1990, 1991 and 1992, the proportion of 1989 properties yielding a return greaterthan or equal to their assumed debt service remained nearly constant.

This result shows how powerful the debt service coverage ratio might be underconditions of long-term commercial leases: if property values drop, rental values remainbasically constant, as long as the mobility of the tenant is blocked through a long-termcontract without the option to move. In most legal systems, with some notableEuropean exceptions, as noted above, such contracts are possible (and typical for themarket).

The simulation analysis also showed the powerful effect of diversification. Alender having a portfolio diversified across a number of vintages (years of origination)fared substantially better than one holding only one vintage (e.g., 1989). This suggeststhe value of having banks in the sector permanently (that is through the cycle) ratherthan forcing them out as conditions deteriorate and letting them back in near the top ofthe cycle.

Valuation Techniques

The key to underwriting using the LTV ratio is valuation. Diverse traditionsand/or regulations govern valuation in different countries. There are no generaltechniques for valuation which are adequate for all purposes. Box 1 gives an overviewover the main methods.


Box 1: Main Property Valuation Methods

• Open Market Value: The Open Market Value is defined as the price at which an asset is tradedbetween sellers and buyers on an open market. In practice, determining the open market value of anasset means drawing inferences from the values obtained in transactions (sales) on properties of aquality and location similar to the property which is to be valued. The open market value has,therefore, comparatively limited information requirements.

The rationale for applying the open market value is information efficiency. It is assumed that marketparticipants are rational, and process all available information in forming future revenue and costexpectations while determining the current market price. As a result, today’s market prices aredeemed to be unrivaled in their accuracy in assessing the value of an asset. The drawback is thatonly a low proportion of commercial real estate is actually turned over at a particular point in time(typically below 2 percent), making the observations of market prices of questionable statisticalsignificance.

• Income Value: Income Value is a method in which discounted cash flows are used to determine thepresent value of the income streams derived from a real estate investment. In practice the methodrequires explicit analysis of all factors determining present and future costs and revenues of theinvestment. Information requirements include, over the entire holding period, an estimate of: the rentstream, the future operating and capital cost stream, and the appropriate discount factor.

Theoretically, discounting all cash flows that the investment yields over the holding period is the mostaccurate approach to property valuation. Assuming rationality and full information on the part of allmarket participants, the open market method would become identical with the income value method.The income value method makes the assumptions explicit, however, and considers a much broaderset of information (for example, rental agreements and expense conditions).

• Reproduction Value. The Reproduction Value determines the cost of reconstructing the building inthe same location with the same characteristics. In practice, the reproduction value requires anassessment of the construction costs necessary to replicate the building. Information requirementsare therefore rather low. Valuing real estate according to a reproduction value starts from anunrealistic assumption, namely, the theoretical possibility of exactly copying the project. By contrast,both open market value and income value include demand aspects, such as expected rental income.

The position taken by the reproduction value method is not as weak as it seems. A closer look atproperty cycles shows that it is typically errors in the assessment of demand factors that makeproperty valuations unreliable and cyclical. In particular, excessive rental growth expectations andthe application of inappropriate discount factors may render valuations inaccurate. Although valuinga property based only on the reproduction value would clearly be inappropriate, valuing a propertywithout considering reproduction costs is hazardous.

Whichever approach is applied, common valuation methods for properties do nottake into account the different credit risk positions of the lender vis-à-vis the equityinvestor. Lenders have limited potential for capital gains, but high potential for capitallosses, while investors typically can take advantage of the full distribution of propertyprices. Equity investors have a “put option,” that is, they can sell the property to themarket at any point in time during their holding period, realizing capital gains. Lendersdo not have this option; in analogy they typically cannot “call” the loan before maturity,


which might be in their interest if property prices decline and the likelihood of a defaultof the investor rises. Because of these serious disadvantages, lenders receive a return:a credit risk premium on the loan and, typically, a preference in the bankruptcyprocedure. Nevertheless, when valuing the collateral, the option characteristics of theirfinancial contract with the investor should be kept in mind.

It is regulatory practice in countries with strong traditions of special regulations inmortgage banking (Austria, Germany, Greece, Denmark) to (i) give preference for anincome value method, and (ii) impose a modification of the cost and revenue factors inorder to arrive at a lender-oriented “mortgageable value.” These regulations typicallyfocus on only the permanent characteristics of the property, i.e., those relevant for along-term investment horizon, which a lender without short-term options needs to take.The following restrictions are applied in the case of Germany:

• A whichever-is-lower principle for rent assumptions, using the lower of, currentcontract or market rents, combined with the requirement to use permanentlyachievable values.

• Assumption of minimum operating costs to arrive at net rents (25 percent of grossrents for residential, 30 percent for commercial property).

• Minimum discount factors (currently 5-5.5 percent for residential, 6-7 percent forcommercial property, adjusted periodically as market interest rates change).

• A general safety margin (10 percent) deducted after the income value has beendetermined.

The major differences between the mortgageable value and open market valueare shown in Table 5. As noted above, the purpose of the two valuations is different.Investors are, quite properly, primarily concerned with likely market value over theinvestment (holding period) time horizon while lenders are concerned with sustainablevalue over the remaining loan maturity. Market comparables are appropriate from theinvestor perspective as he is concerned about near term sale. The mortgageable valuemethod with conservative assumptions about rents and expenses is more suited for thelong-term lender.


Table 5 Property Valuation Techniques: Open Market Value vs. Mortgageable Value

Mortgageable Value Open Market Value

Main Function Protection of creditors “True” economic value forinvestors

Technique

− Approach Discounted cash flow Market comparables

− Discount Factor Refinancing costs of creditor Market yields reflectingalternative investments

− Gross Rent Estimation Project or market rents Project rents with market rentgrowth forecast

− Operating Cost Estimate Minimum percent of rentassumption

Project specific with marketexpenses growth forecast

− Forecast Period Loan period Investment period

Practice

− Determination Bank supervisors Professional organizations

− Enforcement Higher capital risk weightsHigher finance costs

Courts

A major factor in the mortgageable value approach is the requirement to useminimum discount factors. Rather than use current market yields for investors (referredto as capitalization rates) the requirement is to use a discount rate related to therefinancing costs of the lender (in this case the mortgage bank). In the Germanexample this means a discount rate based on long-term rates in the economy (basedon the loan term). Open market valuations often use short-term discount rates. Short-term rates are, however, much more volatile than long-term rates, introducing extravolatility into property valuations. The argumentation transfers to property yield levels:14

lenders should not accept yield levels (high valuations) below their long-term risk-adjusted opportunity (financing) costs. An empirical example for France in Figure 9shows the point.

14 Net rent revenues divided by the capital value.


Box 2: Paris Market

In the “hot” Paris city property market, unlike the French provinces, office yields dropped byearly 1989 as low as 4.5 percent, reflecting the strong capital value (property price)appreciation as actual rent appreciation was relatively unchanged. Investors accepted this lowyield as a consequence of high capital growth expectations, spurred by the dynamics of short-term interest rates that hit their bottom by the end of 1988. At that time, however, 3 monthinterest rates stood around 7.5 percent p.a. and long-term bonds at 8.5 percent p.a.. In fact,provincial locations’ yields matched more or less exactly a long-term bond return, implyinglower capital gains expectations. Under conditions of sustainable mortgageable valuation, aminimum discount factor of 6 percent would have yielded much lower property values forlending purposes in the Paris and vicinity markets than those recorded, starting from the samelevel of net rents. It would have capped the amount of capital gains expectations that thelenders would have been allowed to finance.

Figure 9: Yields on Fully Rented Office Properties in Different French Locations

4.0

5.0

6.0

7.0

8.0

9.0

10.0

80 81 82 83 84 85 86 87 88 89 90 91 92 93 94

Province Locations

Paris - La Defense

Paris

Source: empirica

The mortgageable value concept has to coexist with other, investor-oriented,valuation methods. In cases of strong deviations, for instance between open marketvalue and mortgageable value, a conservative approach for lenders is to adopt themortgageable value approach in determining the loan amount. Imposing amortgageable value is primarily a regulatory instrument that limits lenders’ riskexposure, and makes it visible for supervisors. In Germany it has been shown tostabilize the property price cycles. Its significance will depend on the role of banks aslong-term standing investment financiers. In countries where properties are typically


financed with more than debt, such lender-oriented valuation techniques are less likelyto become a common standard.

COMBINING THE UNDERWRITING FACTORS: RECENT U.S. EXPERIENCE15

Regulatory Background

Historically, commercial banks and savings and loan institutions in the U.S.together provided funding for approximately two-thirds of the commercial mortgagemarket. Today about one-half is still granted by commercial banks but the other half ofcommercial mortgages are packaged into asset-backed securities and sold toinstitutional investors in the form of bonds. This shift reflects both Congressionallymandated restrictions on the provision of commercial real estate credit from federally-insured depositories and development of the securitization funding techniques.

The major recent restrictions on commercial real estate lending passed in recentyears include:

• FIRREA – the Financial Institutions Reform, Recovery and Enforcement Act of1989 restricted the commercial real estate lending authority of a savings and loanfrom 40 percent of assets to four times the institution’s capital. Regulatorysupervisors responsible for resolving savings institution problems invariably citedover-investment in commercial real estate, unsafe and unsound commercial realestate loans, and speculative land loans as leading causes of failure. FIRREAreduced the ability of savings and loans to originate and retain commercialmortgages for permanent portfolio investment.

• FDIC Improvement Act – Credit rating agencies, such as Moody’s InvestorsService and Standard & Poors, downgraded significantly more commercial bankdebt in 1990 and 1991 than was upgraded. Invariably, credit analysts identifiedreal estate problems in banks as the leading cause of the credit deterioration. TheCongress passed the FDIC Improvement Act of 1991 which required Federalregulators to adopt guidelines for banks and savings institutions making real estateloans. The guidelines, which became effective in March 1993, established LTVguidelines that differed by types of real estate collateral. The maximum LTV ratiospermitted for management approved loans included: 65 percent for raw land, 75percent for land development, 80 percent for commercial construction and 85percent for residential construction, improved commercial property and farm land

15 From Handorf, 1997


loans. Exceptions to these guidelines have to be reported to, and approved by, theBoard of Directors of the institutions.

The restrictive legislation and regulatory guidelines are partly responsible forchanging the mix of lenders involved in commercial real estate finance and tighteningunderwriting standards in the 1990s. Savings and loans are no longer majorcommercial real estate lenders. Maximum LTV ratios are much smaller than commonduring the last real estate cycle. The emergence of commercial mortgage-backedsecurities (CMBS) represents the other key change affecting commercial real estatefinance.

Box 3: Commercial Mortgage-Backed Securities

One result of the regulatory initiatives established by Congress and banksupervisors is a shift from crude qualitative rules to prescribed quantitative standards.Future ratings may evolve according to mathematical option pricing models already

Commercial Mortgage-Backed Securities

A commercial mortgage-backed security (CMBS) allows institutional investors to purchase abond backed by a pool of commercial mortgages. Depending on the size of the loans and thepool there may be a few or many loans in the pool. While each CMBS is structureddifferently, most issues share certain similarities and promote standardization and riskdiversification. A lender originates commercial mortgage loans and uses the assignedmortgage liens and leases as collateral to issue a security. The lender assigns the collateralto a trustee for a special purpose vehicle (trust). The trustee receives cash flows from theproperties and/or mortgagors to be distributed to investors according to a priority establishedin the bond indenture. Some CMBS classes are very high quality and carry a “AAA” rating(as assigned by a national rating agency). Bond classes structured to provide less protectionagainst default are assigned lower credit grades. A rating between AAA and BBB qualifies asinvestment grade and is generally eligible for purchase by institutional investors. High gradeCMBS classes are often protected by junior or subordinated classes, cash reserves or over-collateralization. Speculative CMBS classes are assigned a higher risk rating of BB, B orCCC. Rating agencies play a major role in commercial real estate finance by setting ratingsstandards. The ratings are based on LTV and debt-service coverage ratios, prepayment riskand a number of other factors (see Table 7 below). A major difference between CMBS andEuropean mortgage bonds is the use of only commercial real estate collateral (most Danishand German mortgage bonds are backed by mixed pools of commercial and residential realestate loans) and by the use of credit enhancement (senior-subordination, significant over-collateralization). The high credit quality of the European mortgage bonds is based on strictand conservative regulations on the collateral (e.g., 60 percent maximum LTV usingmortgageable value definition) and the strength of the issuer (the bonds are obligations of theissuing bank and thus backed by its capital as well as the collateral).


used to measure and quantify prepayment risk embedded in mortgage loans andMBS.16

Regulatory Approaches

Qualitative Rules: In the U.S. commercial banks and savings institutionshistorically considered a variety of qualitative rules prior to originating and pricingcommercial real estate loans. Table 6 illustrates various criteria often included in loanpolicies developed by depository institutions. Seasoned commercial loans structuredwith low LTV ratios, minor balloon risk, and no recent payment delinquencies wereconsidered to expose lenders to low credit risk.17 Newer, larger, and fully rentedproperties capable of generating high debt-service ratios, constructed with excellentmaterials, well maintained and located in stable or improving geographical areas werealso recognized to provide lenders less exposure to default.

Table 6 Historical Qualitative Underwriting Factors

Factor Low risk Average risk High risk

Loan-to-Value <65% <75% >85%

Debt-Service >1.35 >1.15 <1.05

Delinquency 0 last 36 months 1 last 24 months 1 last 6 months

Occupancy >95% >90% <85%

Age (years) <10 <20 >25

Balloon risk Minor Average Substantial

Construction Quality Excellent Average Poor

Deferred Maintenance None Average Substantial

Neighborhood Excellent Average Poor

Source: Handorf 1997

16 The option pricing approach is described in more detail in Handorf 1997.

17 Balloon loans are loans that amortize over a long time period, for example 20 years, with a requirementthat the outstanding balance be paid off in a lump sum at an earlier time, for example 5 years. Balloon riskrefers to the risk that the borrower may not be able to refinance the loan (from the proceeds of another loan orsale).


The historical rules provided little explanation of terms such as “substantial” or“poor.” Even though qualitative underwriting rules often included numerical thresholdsapplicable to both the LTV and debt-service ratio, the categorization of risk wasunrefined. More precise quantitative standards developed to evaluate the relativecredit-worthiness of CMBS classes differentiate risk much more precisely than thehistorical qualitative rules. The rating agencies conduct computer-based simulationsthat test the stability and consistency of cash flow under different economic andprepayment scenarios. In addition, the thresholds used to quantify risk have becomemore stringent.

Quantitative Standards: The credit rating agencies consider many elements ofrisk prior to assigning a letter grade to an asset-backed security. Credit ratingsassigned by the national agencies range from “AAA or Aaa” to CCC or Caa”. Eachletter grade, except for the highest triple-A rating, is also assigned a number betweenone and three or a sign of plus or minus to further distinguish relative credit risk. Table7 illustrates the quantitative ratios for different property types and ratings. Theassociated probability of default is for corporate bonds of different ratings classes andis shown to demonstrate the association between the probability of default and rating.18

Table 7 Quantitative Standards for CMBS

Credit Rating

AAA AA A BBB BB B CCC

RatioStandards byPropertyType

Office

Debt Service 1.85 1.55 1.45 1.35 1.30 1.25 1.2

LTV (%) 45 55 60 65 75 80 85

Retail

Debt Service 1.8 1.5 1.45 1.35 1.25 1.2 1.15

LTV (%) 50 60 65 70 75 80 85

18 Note that construction and development loans are not rated. They are typically shorter term and riskierloans and not included as CMBS collateral.


Credit Rating

AAA AA A BBB BB B CCC

Apartment

Debt Service 1.75 1.5 1.4 1.3 1.2 1.15 1.1

LTV (%) 50 60 65 70 80 80 85

Probability of default

Annual (%) 0.02 0.19 0.15 0.49 1.76 4.64 6.79

5 year (%) 0.1 0.9 0.8 2.4 8.5 21.1 29.6

Source: Duff & Phelps Credit Rating Co. and “Revisiting the High Yield Market”, Edward Altman, FinancialManagement, Summer 1992

More risky CMBS classes are priced to offer investors higher yields than lessrisky classes. Typically, the yield for each bond is compared to a government Treasurysecurity of comparable duration. As of the mid-1990s, CMBS classes rated “AAA”yielded approximately 85 basis points above comparable maturity Treasury securities,while “BBB” classes yield around 160 basis points above the comparable Treasury.The yield spreads change over the business cycle; spreads increase in an economicrecession when investors become more concerned about default. CMBS pricing alsoshifts in response to the real estate cycle.

The credit rating agencies consider numerous types of additional informationprior to rating and selling a CMBS. The precise amount of over-collateralization orexternal support required to service a security depends on factors additional to theratios listed in Table 7 including:

• Local Market Characteristics: The rating agencies will examine the stability andvacancy/absorption trends within the regional and local markets and compare thebuilding’s physical condition, appraised value, rent rolls, and expenses with otherbuildings within the market area.

• Location and Diversification: Regional economies exert considerable influence onthe value of real estate. Geographic regions in the U.S., such as the Midwest, thatare able to steadily attract a diversified mix of industries are perceived to be lessrisky. Other areas, such as the Northeast or West, that are subject to “boom-to-bust” cycles, dependent on one industry or subject to more “political” risk arebelieved to be more risky. Bonds secured by many properties diversifiedthroughout the country tend to achieve higher credit ratings than securities backedby few properties.

• Property Type: All other things being equal, a CMBS backed by hotel or officeloans will require more collateral, supplemental credit enhancement or better


financial ratio than a security collateralized by multi-family property. Apartmentshave many short-term tenants and the vacancy or bad debt related to one unit haslittle adverse financial effect on the sources of repayment. Hotels, by contrast, incursignificant fixed operating costs and small differences in occupancy rates canproduce large changes in operating cash flow.

• Property Age and Condition: Modest seasoning of a well-maintained officebuilding allows rating agencies to better assess operating history than either an oldor brand new structure. Older buildings may suffer from deferred maintenance orfunctional obsolescence. New buildings may be poorly designed, include untested,defective operating systems.

• Tenant Mix: The number of tenants, space occupied by each tenant, and the creditrecord of important anchor tenants are key factors that affect the likely stability ofcash flow required to service mortgage loans and bond classes. In addition, leaseterms, such as rent escalation clauses, expense stops, renewal and cancellationoptions, and expiration schedules, also affect the stability of cash flow.

• Property Management: Credit risk is affected by the experience and depth ofproperty managers. Successful managers are able to adapt to local economic andcompetitive conditions and minimize tenant turnover and vacancy rates. Insurancemust be sufficient to protect bond investors from loss in the case of propertydamming or interruption of business from natural disaster.

• Agency Relationships: The servicers and trustee appointed to administer a CMBSmust be responsible, experienced, financially capable. The servicer is responsiblefor collecting funds from the mortgagors and distributing the funds to the trustee.The servicer must ensure that taxes, insurance premiums, and assessments arecurrent. The servicer initiates foreclosure proceedings upon the event of default.The trustee holds the mortgage documents on behalf of the investors andperiodically describes the status of the collateral to investors.

• Legal and structural risks: These are risks applicable to the proposed security.For example, does the bond retain sufficient cash reserves? Are there potentialenvironmental liabilities? Are the junior or subordinated bond classes able tosupport the senior classes in the event of a recession?

The preceding factors individually and collectively illustrate the structure, riskand underwriting procedures common to a CMBS. Because approximately one-half ofcommercial mortgages in the U.S. are now financed through the CMBS structure,lenders and investors must be familiar with the standards imposed by the credit ratingsagencies. The quantitative standards expand the number of factors applied, anddifferentiate financial ratios more precisely than the factors previously used bydepository institutions to rank relative credit risk. The regulators also monitor ratings


(both initial and up-grades and down-grades) as evidence of underwriting and servicingperformance of lenders. There has been discussion of using rating agencyassessments to supplement examinations and perhaps to even become the basis forcapital risk weighting.

COMMERCIAL REAL ESTATE LENDING IN POLAND

Overview of Real Estate Markets

After seven years of transformation the commercial real estate market in Polandis still emerging. The largest and most rapidly developing real estate market in Polandis Warsaw. More investment is located in Warsaw than all other Polish cities combined.There is, however, an opinion that during the next few years, the geographicconcentration of development will significantly change. Offices of domestic andinternational companies are beginning to open in larger Polish cities, Poznan inparticular, where the proximity of the border has led many Germany companies to openoffices.

According to brokers, the Warsaw market has experienced excess demand overthe past few years and such agencies estimate that this situation will continue foranother 2 years. This is particularly true for class A offices with both internationalcompanies and, more recently, larger domestic companies seeking to develop theirheadquarters. Newly constructed buildings have been at least 80 percent pre-leasedduring the construction phase and the renting of the remaining 20 percent of the spacehas taken less than two months, although there are some signs of reduced levels ofpre-leasing (down to 50 percent), perhaps reflecting a maturing market. Re-leasedoffices have been also rented within 2 months but more frequently over 2 to 3 weeks.Vacancy rates are currently less than 5 percent in the Warsaw market and are evenlower for higher class offices.

Market studies (typically by less than objective leasing agents) suggest stronggrowth in demand, in excess of 125,000 square meters per year for the next 3 years.19

In part this is due to the lack of high quality (A and B) space (less than 20 percent ofthe market) and the sharp fall-off in quality from A to B and B to C class space.

19 It is estimated that there is between 2.5 and 3 million square meters of total office space in Warsaw, 400-500 thousand in classes A and B. Brokers estimate that there is a significant backlog in demand arising from theincrease in employment in excess of 100,000 over the past few years and the lack of higher quality space. As ofthe summer of 1997 there were at least 15 projects in preparation totalling approximately 500 thousand squaremeters. Retail stock was at least 1.25 million square meters with highly variable rents ($20-80 per square meterper month).


Despite perceptions of excess demand, rents appear to have been stable inWarsaw (typically in USD or DM terms). Rental rates in Downtown or related areas arecurrently:20

• Class A: $36-55 per square meter

• Class B: $26-35 per square meter

• Class C: $15-25 per square meter

Rents are often contractually readjusted on an annual basis by U.S. $ or DMinflation or by a pre-determined increase (often 3 percent). There is evidence ofinstitutions delaying their decisions to rent or purchase an office at these levels as theyhave begun to approach or exceed those in the rest of Europe.

Leases are typically short term with 85 percent of Class A leases having terms of5 years or less (Figure 10). Less is known about Class B lease arrangements. Tenantshave the option to leave earlier than their expected leasing term, usually with aminimum early notice of 3-6 months. The mobility of tenants is quite high, creating arisk for banks with longer term loans (extended vacancy periods, marketing, re-furbishing for new tenants etc.).

Figure 10: Typical Lease Terms

Source: ºaszek

20 Net rents before 22 percent VAT and utilities charges left to tenants (typcially $3-6 per square meter).Cost of common space could add 15-30 percent to rents as well.

A Class Buildings

40%

45%

15%

3 Years 5 Years 7 Years


A potential risk for long-term lending comes from a leveling or decline in rentsonce the initial leasing terms expire, which could occur if development begins todiversify out of Warsaw. This has been the experience of Prague and Budapest, bothof which experienced booms due to a massive entrance of firms seeking high qualityspace, leading to a period of accelerated construction but with supply eventuallyoutstripping demand. However, land supply in Warsaw is subject to significantconstraints, reflecting uncertainties in land titling and a lengthy regulatory processwhich may slow down the ability of the market to meet demand and thereby tend tokeep rents high.

Property development is still uncertain and difficult in Poland for the followingreasons:

• Lack of equity of developers (usually a minimum of 30 percent required by banks)which cannot be funded solely through the donation of land but requires additionalcash (often necessitating a foreign partner).

• Scarce, expensive land, usually under control of local governments with varyingdegrees of sophistication in property development.

• Complex and time consuming permit process for infrastructure and constructionapproval by local authorities.

• Difficulties in meeting bank requirements of minimum pre-leasing before buildingcompletion.

• Inexperienced developers.

• Inexperienced banks (lack of standardized application, review, approval anddisbursement procedures).

Commercial Real Estate Lending

Commercial property lending is quite new in Poland, having developed only inthe past 3 years primarily in the Warsaw market. Currently there are only a few playersin the market, including developers, leasing agents, large city governments, and banks.There are no standardized underwriting procedures, documentation or informationsources and banking regulations are only now being developed.

Only a few banks are active in commercial real estate lending: PBR (historicallythe first and most experienced), PBI, Handlowy and PKO SA. A number of other banksare known to have made some commercial real estate loans, including PBG, PKO BP,PBK, Sl�ski and BPH.


It is generally not possible to identify commercial property loans on bank balancesheets. Such loans may be classified as non-housing real estate, corporate debt, or insome cases equity investments.21 There is no distinction between development orpermanent finance (almost all lending to date has been for development), developmentfor sale or rental, or among property types. Large banks are often asked to provideproject equity as well as debt, as local developers have a shortage of equity capital. Alimited equity position may improve the bank’s control of the project but suchinvestments are considerably more risky and less liquid than loans.

Until recently, most project development and finance was provided by foreigndevelopers with foreign funding. This situation has changed, since large Polish bankswith hard-currency deposits and access to foreign lines of credit have entered themarket along with Polish developers (often subsidiaries of construction companies)often acting in partnership with Polish municipalities (key actors as land owners andpermit providers).

There are no available figures on bank or sector exposure. The NBP obtainscase-by-case information only when asked for special authorization to exceed loan limitratios (10 percent of bank equity for a loan, 15 percent of equity for a consolidatedborrower).22 It is noteworthy that current NBP capital regulations accord a 50 percentrisk weight to commercial property loans (as opposed to 100 percent for residentialproperty loans (this issue will be discussed in more detail below). Bank exposure is stillquite concentrated in the Warsaw market, primarily in hotel and office buildings. Onemajor player, PBR, declared in 1995 its intention to limit its financing of high price officeprojects in Warsaw and to look for projects in other cities in order to diversify itsportfolio.

The average term of a loan is between 5 and 10 years (the range of 2 to 15years) starting typically after a grace period on repayment of capital during theconstruction phase which lasts 18 to 24 months. Interest rates are variable (typically inhard currency terms although there is a growing use of Zloty loans), with USD interestrates of 12 to 20 percent.

21 In the U.S. commercial banks cannot take equity positions in real estate (except for real estate owned asa result of foreclosure and own office real estate). Thrift institutions can invest in real estate through servicecorporations but must hold 100 percent own funds (equity) against the investment. This is also the treatment forGerman mortgage banks.

22 These limits effectively preclude entry by small and medium banks into the sector. Only a few banks (i.e.,those listed above) have the equity to finance class A or B projects and a syndication market has not yetappeared.


The loan-to-value ratio is the key underwriting parameter for banks. The typicalmaximum LTV is 70 percent. Valuation is usually done by outside appraisers who areexpected to use the three usual methods (comparable market price, discounted cashflow and historical cost). However, there is no well identified definition of value in Polandand the information systems necessary to support high quality appraisals using any ofthese models do not exist. Valuation is exceedingly difficult in a transforming marketlike Poland. Comparable sale prices are often difficult to find as many transactions arenot truly “arms length” and subject to various motives (e.g., tax) which may biaspublished information. Rent and expense information is typically readily available, butthe forecasting of these variables and the selection of a discount rate are highlyproblematic in a volatile economy, which also makes cost-based (reproduction)valuations are suspect.

More experienced banks conduct a cash flow analysis and compute a debt-service coverage ratio (DSCR) but there are no standard limits on this variable. Banksattempt to manage credit risk in the commercial market by offering LTV ratios a little lower,say 65 percent to 70 percent, than in developed markets. The DSCR is a bit higher,around 1.2, than that typically required by U.S. banks. Overall, there is little difference inunderwriting standards for commercial real estate in Poland as compared with the U.S.market. Depending on the borrower and the perceived risk of the project, the bank mayask for assignments of rent contracts or payment of rents into an escrow bank account,at least during a limited period.

In summary, the commercial real estate lending market in Poland is in a nascentstage of development. At this point in time it is not likely that any banks have a majorconcentration of this risk (relative to the size of their portfolios and capital). However,the size of many commercial real estate loans is such that a concentration could rapidlydevelop. Thus it is important for regulators to develop adequate monitoring capabilitiesas well as appropriate guidelines for bank involvement in this sector before substantialportfolio accumulation occurs.

LESSONS FOR REGULATION

As the commercial real estate market develops in Poland, it will be important tocreate a strong and stable environment for lenders. This starts with a positive andstable macroeconomic environment. In addition, the market will be well served with animproved information base, a strong legal system supporting creditors and a strong andstable regulatory environment. Development lending has been a key part of nearly allpast commercial property market crises. Oversight and control of developmentactivities as well as promulgation of policies promoting stable property markets canreduce the likelihood that Poland will experience similar problems to those of othercountries in recent years.


Promote Stable Property Markets

A special feature of property markets is that because they represent largely non-tradeable goods, their market prices may be strongly subject to both domesticeconomic conditions and government policies. Political intervention into propertymarkets has been carried out in a number of countries with the intention of stimulatingregional growth, providing jobs and encouraging investment in particular types ofproperties or local markets. The primary tools for intervention in both residential andcommercial markets have been tax incentives and subsidies.23 As the experiences inthe U.S. during the 1980s and Germany in the 1990s have shown, abrupt changes inincentives can lead to sharp price effects. The resultant volatility can be destructive forlenders, particularly those engaged in speculative development finance.

Application of fiscal incentives for property development in major markets suchas Warsaw at the current time should be avoided. Both demand and supply in themarket are subject to major uncertainties that are typical in an emerging market.Demand is based on expectations of future revenues and expenses that are verydifficult to forecast and susceptible to large fluctuation. Land supply is still largelycontrolled by local governments and subject to disparity in the timing and quantity of itsavailability. As the Polish market develops, it will be important to develop land usepolicies that allow supply conditions in major markets to develop steadily. This caninclude regularizing the land development process (including title registration andtransfer) and planning ahead the amount and location of land offered for development,typically a local government responsibility in market economies. Taxation of idleproperty can also discourage speculation. On the demand side, both national and localgovernment should avoid large changes in the fiscal treatment of property developmentthat can lead to unsustainable investment decisions and generate excessive realdemand cycles.

Improve the Information Base

A core lesson about public management of property markets from the diverselending crises is the need to generate financially relevant information for all marketparticipants. In Poland there is a general lack of reliable data on property transactions,prices, characteristics and lease conditions. Market players should be encouraged toinvest in information for both the property and the property finance markets.Information should be generated for internal use by regulators and supervisors and for

23 Classic examples are the British business expansion schemes, which give tax preferences to commercialproperty investments in designated areas, the German reunification policy which for some time was dominatedby tax-stimulated construction-led growth and accelerated tax depreciation of income property in the US in the1980s.


public dissemination to appraisers, investors and lenders. Public dissemination of abroad set of real estate related characteristics will enhance transparency in the marketand guide market players to make better informed decisions. Databases should havethe following characteristics:

• Diversified information base: This requires including a variety of indicators, such asproperty prices, capital appreciation rates, yields, rent data (for both new andexisting contracts), and returns on property investment instruments (such asproperty trusts) simultaneously.

• Diversified information sources: Databases should not rely exclusively oninformation whose generation may be linked to particular economic interests (e.g.,real estate agents’ data). There should be consideration of incentives or initiativesfor data generation and public dissemination by neutral parties.

• Diversified statistical coverage: Record statistical values that reflect the entiredistribution, not just parts (e.g., “top rents”). The public use of price/rent data thatconcentrate on small parts of these distributions should be discouraged as they canpresent a non-representative view of the market.

• Increased sample survey bases: Publicly-managed surveys (for example, inconnection with a property taxation system) can be a valuable source of informationfor the market.

Given the public good aspects of this information and the need to becomprehensive and unbiased, it is recommended that a government body beestablished to manage such a database. Donor seed funding for such a body would bea good use of resources, given the interest in this sector.

Strengthen the Legal System

Insufficiencies in legal systems have exacerbated many commercial lendingcrises, reducing even further the value of already questionable collateral to the lenderin repossession of the property. When lender rights to collateral are constrained ornon-existent, the commercial mortgage credit market will either develop discontinuously(increasing the damaging potential of big bang effects once the rules are changed), orin forms that create even more lending risks (e.g., uncollateralized loans).Furthermore, the value of the commercial mortgage collateral may suffer if key legalasymmetries are present between landlord and the tenant.

In Poland, it is important to strengthen the legal system’s protection of lenderrights in order to discourage commercial mortgage default. In particular, statutory liensthat discourage mortgage lending should be eliminated. Mortgage lenders should havea priority over other creditors in bankruptcy procedures opened on commercial


borrowers. Care should be taken to avoid onerous consumer protection elements inlegislation governing commercial mortgage lending (e.g., rent ceilings, options fortenants to break lease contracts).

Improve Bank Regulation and Supervision

Capital Treatment: An important instrument of regulatory policy is risk-basedcapital requirements. The traditional view expressed in banking regulation has been auniform (“brush”) approach to capital adequacy, with few deviations from broad classesof assets. 24 Uniform capital holding requirements (the principle of the Basle guidelines)do not observe any risk-premia relationships.25 As a result the uniform approach issubject to adverse selection problems that occur when different risk classes are pooledinto one. For instance pooling development lending (“property lending”) and long-termstanding investment finance of commercial real estate into one class of capitalrequirements means creating incentives for the lender to do more development lending,the higher risk class.

An alternative and emerging approach is to develop simple rules matching creditrisk content and capital requirements. By virtue of its numerous derogations, Basle hasde-facto already moved from a uniform (brush) capital holding approach to a simple-rules approach. The European Commission has proposed to convert the derogations ofArticle 11 (4) Solvency Directive allowing a 50 percent weight for low risk commercialreal estate loans into a permanent rule (subject to additional requirements).26 TheReserve Bank of Australia applies a 100 percent risk-weighting to residential mortgageloans with valuation ratio of 80 percent and above. The U.S. has differentiated rules formulti-family mortgage lending by loan-to-value and debt service coverage ratio. In thepractice of determining such rules, however, there is generally an absence of statisticaltesting and periodic revision of the risk-capital holding relationships. However, it isclear that the market is moving in this direction as the discussion of the ratingsapproach for CMBS above suggests.

In Poland, the current risk based capital requirement for commercial real estateloans is 50 percent whereas the requirement for residential mortgage loans is 100percent. This treatment is based in part on the perceived weaker position of the lenderin residential foreclosure proceedings. There is no differentiation between loans for

24 Examples are the treatment of residential mortgage loans (unconditional 50 percent risk-weighting) andgovernment loans (unconditional 0 percent risk-weighting).

25 The Basle guideleines are recommendations for minimum risk-based capital requirements for banks.

26 For loans on leased buildings with LTV less than or equal to 60 percent and debt-service coverage ratio inexcess of 1.2.


development or standing investment. Although regulatory standards are movingtowards differentiation of capital requirements by perceived risk, given the lack ofexperience with this type of lending, the potential for volatile property price movementsand the lack of information on commercial property characteristics and data in Poland,it is recommended that the NBP increase the commercial mortgage risk based capitalrequirement to 100 percent, on par with residential loans. As experience with andinformation on commercial mortgage lending is gained, the NBP could consideradopting lower risk-based capital requirements for certain conservatively underwrittenstanding investment loans.

The regulatory treatment could be strengthened to encourage conservative loanunderwriting along the following lines:

• Discriminate regulatory treatment between long-term standing investmentloans and development loans (property lending), for example throughrestrictions on or higher capital requirements for development lending.

• Provide incentives for higher investor equity, both on the balance sheet and inthe project (that is, the total loan-to-value of the project). Finance theory andempirical evidence both suggest that high investor equity is the strongest deterrentagainst default, depending on the strength of bankruptcy legislation. Lenders shouldbe required to obtain and assess project and borrower equity.

• Provide incentives for loans given under a favorable risk position for thelender in the case bankruptcy (for example, start with lower capital requirements forfirst as opposed to second mortgages, or senior as opposed to subordinate loans).

• Provide incentives for loans based on sufficiently documented incomegeneration characteristics that support debt service. Assessment should bebased on documented (pre-) lease contracts over a minimum number (for example5) of years. Consider defining a debt service coverage ratio for regulatory treatment(for example 120 percent of debt service covered by net rent revenues).

Valuation: The discussion above emphasized the importance of valuationtechniques. It is important to establish property valuation techniques that take intoaccount the asymmetric risk position of the lender as opposed to the equity investor inthe project. The open market value provides an estimate as to what the project couldsell for today and the mortgageable value provides estimate of the long termsustainable value. For commercial real estate lending, it is recommended that:

• For bank underwriting purposes, the use of cash-flow oriented valuationmethods that utilize information about expected rental income and costs indescribing the income-generating capacities of the project should be required(along with open market valuation methods).


• Standards for cash flow valuations of projects be developed. Such valuationsshould be based on a long-term time horizon (at least the term of the loan) forforecasting income and expense and use of a discount factor (e.g., based ongovernment bond yields)27 that reflects the term. In addition, cash-flow orientedvaluations should concentrate on reasonably expected permanent characteristics ofthe project, in particular rental income.

Regulation: Mortgage lending in general and commercial real estate lending inparticular is a complex business. There is an advantage to specialization in mortgagelending issues for bank supervisors. The U.S. Office for Housing Enterprises Oversight,or the German Treuhänder acting for mortgage banks in the Federal Banking OversightOffice, are examples. Even without special agencies, a minimum competency in theoversight of mortgage and commercial real estate lending should be established withingeneral (e.g., banking) regulatory agencies.

Although it is essential for regulators to monitor lenders’ real estate exposure,and issue warnings regarding unsafe or unsound practices, it is important not to createregulatory stop-and-go policy, for example through unsustainable portfolio regulation orsystematic policy bias against commercial real estate. This should be the means of lastresort in the face of crisis and may, if practiced continually, become part of the problemrather than the solution. Excessive regulation, if permanent, may discourage thespecialization of lenders in real estate and the development of better risk managementpolicies that can result.28

As an important complement to regulation, industry should have incentives toself-supervise: e.g., by developing common securitization standards (mortgage bond orMBS) that involve credit risk protection measures.

27 Given the short-term nature of the Polish bond market, the use of long-term discount rates would have tobe phased in as benchmarks become available.

28 The empirica study showed with the London example that vintage diversification of commercial propertyholdings may be a stronger hedge against commercial property risk than asset diversification. Also, humancapital accumulation is necessary in order to create internal monitoring capacities of borrowers, externalappraisers, etc.


A related regulatory area is regulation and supervision of securities issuance.Information requirements are the key to giving small-scale investors in Real EstateInvestment Trusts or similar mutual funds sufficient control over their investment.29

In conclusion, commercial real estate lending is an important source of capitalfor the property market. Much has been learned about the role of property lending inreal estate cycles and macroeconomic instability. There has also been considerableexperience as to appropriate underwriting techniques, information disclosure andregulation to manage the risk of this lending. Poland has the opportunity to learn fromthis experience in shaping the industry in the years to come.

29 Experience with real estate investment trusts in East Germany have shown that many small-scaleinvestors were lured into commercial real estate investment schemes by the withholding of key informationregarding concrete projects and overselling the tax advantage argument. The tax savings motive has in the EastGerman case helped to create pools of hot money that might have been avoided with better informationrequirements.

REFERENCES

BANK FOR INTERNATIONAL SETTLEMENTS, Annual Reports 1991-1996, Basle.

Board of Governors of the Federal Reserve System. 1996, “Federal Reserve Bulletin”,Washington, DC.

CLAURETIE, TERRENCE AND G. STACY SIRMANS, 1996, Real Estate Finance: Theory andPractice (2nd edition). Prentice Hall, Upper Saddle River, NJ.

COMMISSION OF THE EUROPEAN UNION, 1997, “Supplementary Pensions in the SingleMarket, a Green Paper,” DG XV: Brussels.

DAVIS, E.P., 1993, “Bank Credit Risk.” Bank of England: London.

DÜBEL, A., U. PFEIFFER, WITH A. KEY, M. LEA AND I. MORLING, 1995, “Risk-based CapitalRequirement and Commercial Mortgage Credit Risk,” Schriftenreihe desVerbandes Deutscher Hypothekenbanken, Fritz Knapp Verlag: Frankfurt.

European Mortgage Federation, 1996, “Hypostat,” Brussels.

Federal Deposit Insurance Corporation, 1995, “Report on the Condition of Banks,”Washington, DC.

GOLDSTEIN, M., AND P. TURNER, 1996, “Banking Crisis in Emerging Economies: Originsand Policy Options,” in: BIS Economic Papers, No. 46, Bank for InternationalSettlements: Basle.

HANDORF, W., “The Risks of Commercial Real Estate Lending,” Real Estate Review,Summer 1997.

HONOHAN, P., 1997, “Banking System Failures in Developing and Transition Countries:Diagnosis and Prediction,” in: BIS Working Papers, No. 39, Bank forInternational Settlements: Basle.

KERL, V., 1994, “Bankaufsichtliche Anforderungen an den Realkredit,” VerlagC.H.Beck: München.

PEEK, J. AND E ROSENGREN, 1994, “Bank Real Estate Lending and the New EnglandCapital Crunch,” Journal of the American Real Estate and Urban EconomicsAssociation, Vol 22.

RENAUD, B, 1997, “The 1985-1994 Global Real Estate Cycle: An Overview,” Journal ofReal Estate Literature, 5: 13-44.


SCHINASI, G. AND M. HARGRAVES, 1993, “Boom and Bust in Asset Markets in the 1980's:Causes and Consequences,” in: Staff Studies for the World Economic Outlook,International Monetary Fund: Washington, DC.

VANDELL, K.D. BARNES, W. HARTZELL, D. KRAFT, D. AND WENDT, W., 1993, “CommercialMortgage Defaults: Proportional Hazards Estimation Using Individual LoanHistories,” in: Journal of the American Real Estate and Urban EconomicsAssociation 4: 451-480.

VAN ORDER, R., 1990, “The Hazards of Default,” in: Secondary Mortgage Markets, 7, 3Freddie Mac, Washington DC.

DATA ANNEX

Figure 11: Germany: "Top Rents" and Broader Sample Measures in Two German Cities

0

10

20

30

40

50

60

70

80

90

100

1980 1985 1990 1991 1992 1993

DM

per

SQ

M

Hamburg: Average Rent Good Quality

Hamburg: JLW-"Top Rents"

Frankfurt: Average Rent Good Quality

Frankfurt: JLW-"Top Rents"

Source: empirica

Figure 12: United Kingdom: House Prices and Office Capital Values in London, 1980-1994

40

60

80

100

120

140

160

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

Inde

x M

ean

1980

/94=

100

IPD London Office CapitalValues

Nationw ide GreaterLondon House Prices

Source: empirica


Figure 13: France: Apartment and New Office Sales Prices in Paris, 1980-1994

0

20

40

60

80

100

120

140

160

180

200

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

Inde

x m

ean

1990

/94=

100

New Office Sales Prices ParisCentre (Bourdais)

Old Apartments Sales Price Paris(Chambre interdepartementales desnotaires de Paris)

Source: empirica

the risks of commercial real estate lendingpdf.usaid.gov/pdf_docs/pnacj140.pdf · impact of the...

Documents